www.nature.com/scientificreports

OPEN

received: 30 June 2015 accepted: 24 November 2015 Published: 22 December 2015

Characterization of marine diatom-infecting virus promoters in the model diatom Phaeodactylum tricornutum Takashi Kadono1, Arisa Miyagawa-Yamaguchi1,†, Nozomu Kira2, Yuji Tomaru3, Takuma Okami1, Takamichi Yoshimatsu1, Liyuan Hou4, Takeshi Ohama4, Kazunari Fukunaga1, Masanori Okauchi5, Haruo Yamaguchi1, Kohei Ohnishi6, Angela Falciatore7 & Masao Adachi1 Viruses are considered key players in phytoplankton population control in oceans. However, mechanisms that control viral gene expression in prominent microalgae such as diatoms remain largely unknown. In this study, potential promoter regions isolated from several marine diatom-infecting viruses (DIVs) were linked to the egfp reporter gene and transformed into the Pennales diatom Phaeodactylum tricornutum. We analysed their activity in cells grown under different conditions. Compared to diatom endogenous promoters, novel DIV promoter (ClP1) mediated a significantly higher degree of reporter transcription and translation. Stable expression levels were observed in transformants grown under both light and dark conditions, and high levels of expression were reported in cells in the stationary phase compared to the exponential phase of growth. Conserved motifs in the sequence of DIV promoters were also found. These results allow the identification of novel regulatory regions that drive DIV gene expression and further examinations of the mechanisms that control virusmediated bloom control in diatoms. Moreover, the identified ClP1 promoter can serve as a novel tool for metabolic engineering of diatoms. This is the first report describing a promoter of DIVs that may be of use in basic and applied diatom research. Diatoms are one of the most effective and diverse unicellular photosynthetic eukaryotes, with perhaps as many as 200,000 extant species found in aquatic environments1. There is broad interest in diatoms due to their substantial contributions to global carbon cycling and oxygen production2, their complex evolutionary background as secondary endosymbionts3, and their unique capacities to produce silica-based cell walls4. Moreover, diatoms present great potential as a source of beneficial chemicals for use in human activities because they produce biofuel precursors such as fatty acids and hydrocarbons that may prove useful in solving ecological problems such as the energy crisis5. To understand diatom biology and the means of diatom exploitation in biotechnology, novel genomic resources and molecular tools have been developed in recent years. Databases of genome sequences and the expressed sequence tag (EST) of the model diatoms Pennales Phaeodactylum tricornutum and Centrics Thalassiosira pseudonana have been released for public use, allowing the identification of distinct metabolic features in diatoms6–9. Recently, the genomic sequences of other diatoms such as Pseudo-nitzschia multiseries (http://genome.jgi-psf.org/ Psemu1/Psemu1.home.html) and Fragilariopsis cylindrus (http://genome.jgi-psf.org/Fracy1/Fracy1.home.html) 1 Laboratory of Aquatic Environmental Science (LAQUES), Faculty of Agriculture, Kochi University, Otsu-200, Monobe, Nankoku, Kochi 783-8502, Japan. 2The United Graduate School of Agricultural Sciences, Ehime University, 3-5-7 Tarumi, Matsuyama, Ehime, 790-8566 Japan. 3National Research Institute of Fisheries and Environment of Inland Sea, Fisheries Research Agency, 2-17-5 Maruishi, Hatsukaichi, Hiroshima 739-0452, Japan. 4School of Environmental Science and Engineering, Kochi University of Technology, Tosayamada, Kami, Kochi 782-8502, Japan. 5National Research Institute of Aquaculture, Fisheries Research Agency, 422-1 Nakatsuhamaura, Minami-ise, Mie 516-0193, Japan. 6Research Institute of Molecular Genetics, Kochi University, Otsu-200, Nankoku, Kochi 7838502, Japan. 7Sorbonne Universités, UPMC Univ Paris 06, Institut de Biologie Paris-Seine, UMR 7238, F-75006 Paris, France; CNRS, UMR 7238, F-75006 Paris, France. †Present address: Center for Innovative and Translational Medicine, Kochi University Medical School, Kohasu, Oko-cho, Nankoku, Kochi 783-8505, Japan. Correspondence and requests for materials should be addressed to M.A. (email: [email protected])

Scientific Reports | 5:18708 | DOI: 10.1038/srep18708

1

www.nature.com/scientificreports/ have also been made available on the U.S. Department of Energy Joint Genome Institute’s website (http://jgi.doe. gov/). In addition, methods of genetic transformation via biolistic bombardment and/or electroporation have been developed for several diatom species including Pennales P. tricornutum10–16, Cylindrotheca fusiformis17,18, Navicula saprophila19, Pseudo-nitzschia arenysensis20, Pseudo-nitzschia multistriata20, Centrics Cyclotella cryptica19, Thalassiosira pseudonana21, Thalassiosira weissflogii22, Chaetoceros sp.23, Cha. gracilis24, and Fistulifera sp.25. Moreover, episomal vector technology based on yeast-derived sequences has recently been developed for P. tricornutum and Thalassiosira pseudonana26. However, many of the processes that control diatom biology and growth in marine environments remain to be elucidated. Recently, it has been proposed that marine viruses play a key role in controlling the blooms of diatom populations27,28. Viruses may also influence the compositions of marine communities and may act as a major force behind biogeochemical cycles29,30. Thus far, several marine diatom-infecting DNA/RNA viruses that cause the lysis of host diatoms have been isolated from both Centrics and Pennales diatoms28, for example, the Cha. debilis-infecting DNA virus (CdebDNAV)31, the Cha. lorenzianus-infecting DNA virus (ClorDNAV)32, and the Thalassionema nitzschioides-infecting DNA virus (TnitDNAV)33. In silico analysis of these genomes has revealed the presence of putative open reading frames (ORFs) such as the replication-associated protein (VP3) gene, the structural protein (VP2) gene, and genes of unknown function28,32. To determine the contributions of marine diatom-infecting viruses (DIVs) to phytoplankton biology and ecology, it is important to further examine the DIV genome and to identify mechanisms that control viral gene expression through the characterization of viral promoters. Moreover, DIV promoter characterization may also advance the manipulation and modulation of gene expression in diatoms. Thus far, this strategy is used primarily through endogenous promoters such as the gene-encoding fucoxanthin chlorophyll a/c-binding protein (FCP) gene (fcp, now called Lhcf), which encodes members of the light harvesting complex superfamily10,18,21,24, frustulin17, nitrate reductase (NR)18,21,24, acetyl-CoA acetyltransferase24, acetyl-CoA carboxylase19, β -carbonic anhydrase 1 (CA1)34, long-chain fatty acyl-CoA synthetase24, β -tubulin24, ATP synthase subunit C24, and elongation factor 2 (EF2)16. Of the endogenous promoters, the fcp promoter has been frequently used in biotechnological applications involving diatoms35–39. However, reports also suggest that fcp promoter activity may not be strong enough to overexpress introduced genes and may promote significant increases in target products35,39. Other studies report that transgene expression follows a circadian pattern due to the presence of light-responsive cis-regulatory elements in the fcp promoter16,40,41. In addition, in some cases, diatom endogenous gene promoters drive species-specific transgene expression and do not function in different diatom species18,20,22. These issues considerably limit the applications of diatom transformation techniques, potentially reducing the utility of diatoms in basic and applied research. To address these problems, the development of promoters that stably induce high-level gene expression is anticipated, which should result in the creation of large accumulations of introduced gene transcripts and proteins. Promoters from viruses that infect plants or mammals have generally been used to transform a wide range of higher plants and mammals, allowing the constitutive and strong expression of introduced genes. For example, the cauliflower mosaic virus 35S (CaMV 35S) and cytomegalovirus (CMV) promoters are known to efficiently facilitate plant and mammalian transformations, respectively42,43. However, only a few reports focus on the transformation of diatoms through the use of CaMV 35S and CMV promoters19,44 in P. tricornutum, and the same promoters do not seem to be effective in other species, such as the Centrics diatom Cyc. cryptica19. DIV promoters may enable us to solve the above-described limitations in the bioengineering of diatoms as well as to infer the ecological roles of DIVs in marine ecosystems. Thus, in this study, potential promoter regions located upstream of ORFs, such as the replication-associated protein (VP3) gene and the structural protein (VP2) gene in the DIV genome (Fig. 1a), were isolated. DIV promoter-driving gene expression capacities in the Pennales diatom P. tricornutum were evaluated and compared to those of the endogenous fcp promoter (Fig. 1b). We then examined the stability of DIV promoters in cells that were cultured under various growth conditions. The effects of the growth stage, the nutrient concentrations in media, and the photoperiods on viral promoter activity were investigated. The application of viral promoters to a Centrics diatom and to unicellular green alga was also examined. We then discussed the structure of a DIV promoter to show how gene expression mechanisms may play a key ecological role in the control of diatom blooms and how the DIV promoter may be used in applied research on diatoms.

Results

Isolation of potential DIV promoter regions.  Potential DIV promoter regions were determined from the

genome sequences of CdebDNAV and TnitDNAV via in silico analyses. The ORFs of the two DIVs were identified using an ORF finder (National Center for Biotechnology Information, http://www.ncbi.nlm.nih.gov/gorf/). Three (VP1, VP2, and VP3) and two (VP2 and VP3) ORFs were identified in the CdebDNAV and TnitDNAV genomes, respectively. In regard to ClorDNAV, the four ORFs (VP1, VP2, VP3, and VP4) were previously described by Tomaru et al.32 (Fig. 1a). Regions (approximately 500 bases long) upstream of the translation initiation codon (ATG) of a putative ORF in the DIV genome were predicted to be potential promoter regions (Table 1, Fig. 1a, and Supplementary Fig. 1).

Construction of transformation vectors.  The potential promoter region of each ORF was amplified via

polymerase chain reaction (PCR) using DIV genomic DNA and specific primers (Supplementary Table 1). In addition to the potential DIV promoters, the endogenous fcp promoter of P. tricornutum and some extrinsic promoters, such as CaMV 35S, CMV, and the nopaline synthase gene (nos) promoter of Agrobacterium tumefaciens, were amplified from the genomic DNA of P. tricornutum and from commercially available vectors, respectively, via PCR amplification (Supplementary Table 1). To investigate the activity of each tested promoter, we constructed a double-cassette transformation vector (Fig. 1a) that contained the enhanced green fluorescence protein (eGFP) gene (egfp) driven by each tested promoter and the antibiotic-resistant gene Sh ble driven by the Cyl. fusiformis FCP Scientific Reports | 5:18708 | DOI: 10.1038/srep18708

2

www.nature.com/scientificreports/

Figure 1.  Schematic diagram for the evaluation of promoter activity. (a) Outline of the construction of transformation vectors and transformations. After predicting putative ORF positions32, upstream regions of the ORFs were determined as potential promoter regions. Potential promoter regions amplified by PCR were used to construct the transformation vectors. The double-cassette vector containing the reporter gene egfp driven by each tested promoter and the antibiotic-resistant gene Sh ble driven by the promoter region of the fucoxanthin chlorophyll a/c-binding protein (FCP) A-1A gene derived from Cyl. fusiformis (termed CffcpA pro.) were constructed. (b) Assessment of promoter activity. Promoter activity was determined by averaging the ratios of egfp mRNA transcript levels to those of Sh ble mRNA transcripts in ten transformants to minimize the effects of copy numbers on the expression of transgenes. These transformants were also used to investigate eGFP protein expression patterns. CffcpA ter.: terminator region of the FCP A-1A gene derived from Cyl. fusiformis. The structure of the ClorDNA genome was modified from Tomaru et al.32. *For the transformation vector of the nitrate reductase gene promoter, we used pNICgfp18 (Supplementary Fig. 5a).

A-1A gene promoter (termed CffcpA pro.). We also constructed a single-cassette transformation vector containing Sh ble driven by each tested promoter (Supplementary Fig. 2) and nat driven by DIV promoters, such as CdP1 and ClP1 (Supplementary Fig. 3), to investigate transformation efficiency. Scientific Reports | 5:18708 | DOI: 10.1038/srep18708

3

www.nature.com/scientificreports/

Source organism/virus

Promoter associated gene

Amplified size (bp)

Ref.

CdP1

Chaetoceros debilis-infecting DNA virus (CdebDNAV)

Putative replication-associated protein (VP3) gene

477

31

ClP1

Chaetoceros lorenzianus-infecting DNA virus (ClorDNAV)

Putative replication-associated protein (VP3) gene

502

32

ClP2

Chaetoceros lorenzianus-infecting DNA virus (ClorDNAV)

Putative structural protein (VP2) gene

474

32

TnP1

Thalassionema nitzschioides-infecting DNA virus (TnitDNAV)

Putative replication-associated protein (VP3) gene

424

33

TnP2

Thalassionema nitzschioides-infecting DNA virus (TnitDNAV)

Putative structural protein (VP2) gene

424

33

Phaeodactylum tricornutum

Fucoxanthin chlorophyll a/c-binding protein A gene

444

10

Cfnr pro.

Cylindrotheca fusiformis

Nitrate reductase gene

774

18

CaMV 35S pro.

Cauliflower mosaic virus

35S gene

454

44,81

Cytomegalovirus

Immediate early promoter regulatory gene

742

44,82

Agrobacterium tumefaciens

Nopaline synthase gene

338

81

Promoter

PtfcpA pro.

CMV pro. nos pro.

Table 1.  Promoters used in this study.

Confirmation of the introduced gene and promoter in the transformants.  Colonies were obtained

from a solid f/2 medium containing 500 μ g ml−1 antibiotics after its transformation via microparticle bombardment using the double-cassette transformation vectors. To confirm the introduction of the two cassettes, a genomic PCR analysis was performed using different primer sets (Supplementary Tables 1 and 2 and Supplementary Figs 4a and 5a) and using the colony-forming cells as templates. We obtained more than ten transformants with the expected DNA fragment length and containing egfp with the tested promoter, with the exception of transformants that were transfected with pNICgfp18. Six transformants that had been transfected with pNICgfp18 possessed DNA fragments of the expected length, including egfp under the control of the NR gene (nr) promoter from Cyl. fusiformis (termed Cfnr pro.) and the Sh ble gene under the CffcpA pro. A typical electrophoretogram of the amplicons in each transformant is shown in Supplementary Figs 4b and 5b. Ten transformants with each of the tested promoters and six transformants with Cfnr pro. were analysed further.

Quantitative analysis of promoter activity.  Variations in the levels of egfp mRNA transcripts normalized

to the internal standard gene (ribosomal protein small subunit 30S gene, rps) mRNA transcripts were detected using quantitative reverse transcription-PCR (qRT-PCR) in 6− 10 transformants with the egfp driven by each of the tested promoters (Supplementary Fig. 6). Variations in the levels of Sh ble mRNA transcripts driven by the diatom endogenous promoter (CffcpA pro.) that had been normalized to rps were also found amongst the 6− 10 transformants (Supplementary Fig. 6). Before determining the activity of each promoter, the egfp mRNA transcript levels in each transformant were divided by those of the Sh ble mRNA transcripts. This method has been used to compare gene expression levels in independent lines and to prevent misinterpretation due to the presence of different transgene copy numbers and possible site integration effects. Figure 2 shows averages and ranges of promoter activity in the 6− 10 independent transformants with egfp driven by each tested promoter. We compared the activity of each tested promoter with that of an endogenous promoter PtfcpA pro. because the PtfcpA pro. has generally been used for the transformation of the marine diatom P. tricornutum10,11,13,39. Average promoter activities in the transformants with Cfnr pro. (P =  0.010), TnP1 (P =  0.012), TnP2 (P =  0.016), the CaMV 35S promoter (P =  0.014), the CMV promoter (P =  0.021), and the nos promoter (P =  0.034) (average: 0.0833− 0.499) were found to be significantly lower than those in the transformants with PtfcpA pro. (average: 2.08, range: 0.0296− 5.97) (Fig. 2). By contrast, the average promoter activity in ClP1 (average: 10.3, range: 2.06− 25.1, P =  0.0058) was found to be approximately five times higher than that in PtfcpA pro. (Fig. 2). The average CdP1 (average: 1.99, range: 0.0917− 7.34, P =  0.93) and ClP2 activity (average: 3.09, range: 0.352− 21.4, P =  0.64) levels showed no significant differences from the average PtfcpA pro. activity level (Fig. 2). The maximum activity levels in the ten transformants of CdP1, ClP1, and ClP2 were higher than that found in the transformants with PtfcpA pro. (Fig. 2).

Effect of culture conditions on DIV promoter activity.  Promoter activities are known to respond to

changes in environmental and intracellular conditions. To explore the effects of culture conditions and growth phases on the activity of the DIV promoter (ClP1) that showed the highest levels of activity (Fig. 2), the abundance of egfp mRNA levels in two independent ClP1 transformants was analysed using qRT-PCR under various culture conditions (Fig. 3). Under a low nutrient culture condition (f/10 medium), the abundance of egfp mRNA transcripts driven by ClP1 was almost identical to that reported for the standard nutrient culture condition (f/2 medium) (Fig. 3b). The abundance of egfp mRNA transcripts in the stationary phase seemed to be higher than that found for the log phase under both low and standard nutrient culture conditions (Fig. 3b). ClP1 can effectively drive the expression of egfp mRNA transcripts in both light and dark conditions during the stationary phase (Fig. 3b).

Levels of eGFP fluorescence.  The fluorescence of the eGFP reporter in the different transgenic lines

described above was quantified via flow cytometry. Figure 4 shows normalized eGFP fluorescence levels for cell sizes in the 9− 10 independent transformants with egfp driven by each tested promoter. The average eGFP fluorescence levels in the transformants containing TnP1 (P =  0.0071), TnP2 (P =  0.012), the CaMV 35S promoter (P =  0.0063), the CMV promoter (P =  0.021), and the nos promoter (P =  0.012) (average: 2.53− 3.24) were found to be significantly lower than those of the transformants containing PtfcpA pro. (average: 6.10, range: 2.21− 11.6) (Fig. 4). The statistical analysis results show that the average eGFP fluorescence levels in the transformants containing ClP1

Scientific Reports | 5:18708 | DOI: 10.1038/srep18708

4

www.nature.com/scientificreports/

Figure 2.  Relative activities of various promoters including DIV promoters in transformants. Ten independent transformants for each promoter were analysed. The promoter activity levels were determined by dividing egfp mRNA levels by Sh ble mRNA levels. The circles indicate the mean triplicate measurements of the independent transformants. The diamonds denote the average values of the six transformants from Cfnr pro. or of the ten transformants from the other promoters. Pro.-less denotes cases in which no promoter was linked to the egfp gene in the transformation vector (negative control). PtfcpA pro. shows the positive control in the transformation vector egfp gene driven by PtfcpA pro. Asterisks indicate the presence of a statistically significant difference derived from the PtfcpA pro. transformants (**P 

Characterization of marine diatom-infecting virus promoters in the model diatom Phaeodactylum tricornutum.

Viruses are considered key players in phytoplankton population control in oceans. However, mechanisms that control viral gene expression in prominent ...
NAN Sizes 1 Downloads 10 Views